Working Paper
Variability can be an important strategic variable in a contest. In a single-round contest, a contestant should choose the most (least) variable performance distribution when the proportion of winners p is low (high).
The authors study a multiple-round contest with opportunities to modify variability each round, based on cumulative performance level x. If only one contestant can modify variability, her probability of winning can increase considerably, with the benefits highest when p is near one-half. High (low) variability is optimal for low (high) x, and there is a range of values of x for which intermediate variability levels are optimal.
If all contestants can modify variability, a similar strategy with the intermediate range shifted to the right is a symmetric Nash equilibrium.
Faculty
Professor of Decision Sciences
Professor of Decision Sciences